NIA Pilot Program PAR-02-049, Subs. 20, Racial/Ethnic Differences End-of-life decision-making is an important aspect of providing quality healthcare, especially for the elderly population. Increasingly, the appropriateness of many of these decisions is being questioned. Studies have demonstrated that many invasive procedures done in seriously ill patients do not significantly alter their course, many patients die without having severe pain or other symptoms addressed, and families felt dissatisfied with their care. Additionally, there are striking racial/ethnic disparities in end-of-life care. Contrary to the usually observed reduced use of procedures for African Americans, the minority group for which the most data exist, in end-of-life care procedures are used at a much higher rate in African Americans than in whites. Given that the medical-technical orientation of care at the end-of-life has been severely criticized and is considered as 'poor' quality of care by some, these observed disparities may reflect yet another example of 'worse' care for minorities. Alternatively, it could represent true cultural/ethnic/racial differences in decision-making for end-of-life care. To address this gap in knowledge and gain further insight of the decision-making process, we propose a qualitative study with the Specific Aims: 1) To explore qualitatively how end-of-life decisions are made, we will conduct focus group interviews with the main participants of the decision-making process for end-of-life care, stratified by race, to assess values, concerns and beliefs that guide patients', surrogates' and physicians' end-of-life decision-making, with a special emphasis on cultural/racial/ethnic differences. 2) To create a comprehensive, culturally sensitive Values Inventory that will incorporate data empirically derived from patients' and surrogates' focus groups with the overall goal to significantly improve the quality of the decision-making process at the end-of life.The proposed study will result in a Values Inventory that will guide physicians' assessments of patients' and surrogates' values and preferences for end-of-life care. This proposed project will lead directly to a RO1 application with the specific aim to assess the feasability and efficacy of using the Values Inventory in a clinical trial.